JP4826362B2 - Method for forming a microlens - Google Patents

Description

  The present invention relates to a technique for forming a microlens used as an on-chip lens or the like such as a CCD solid-state imaging device or a liquid crystal display device.

  In CCD solid-state image sensors and MOS-type solid-state image sensors, in order to increase the amount of light incident on the pixels and improve the sensitivity, a microlens is formed to increase the degree of light collection on the photosensitive portion. The microlenses corresponding to are arranged in a matrix, for example. In order to increase the sensitivity of the CCD or CMOS sensor, it is required to increase the amount of light at the focal point by increasing the area of the microlens. Therefore, it is necessary to narrow the interval between the adjacent microlenses, and specifically, as shown in FIG. 16, the distance between the microlenses 100 arranged in the vertical or horizontal direction is at a diagonal position with respect to each other. It is necessary to reduce or eliminate the separation distance D2 of the microlens 100.

  Such a microlens 100 has a wavelength region with excellent transparency and a region where light can be condensed depending on the material. As a lens material, an inorganic material such as a silicon nitride film or a silicon oxide film is used in addition to an organic material depending on the application. It is preferable that the material can be freely selected. By the way, in order to form the microlens 100, for example, as shown in FIG. 17A, a lower layer portion 101 where a photosensitive portion or a conductive film is formed, a lens material layer 102, and a mask layer 103 made of a resist film. However, a semiconductor wafer (hereinafter referred to as “wafer”) W stacked in this order from the lower side is used. Then, the mask layer 103 is formed in a lens shape as shown in the figure, and the mask layer 103 and the lens material layer 102 are etched by plasma of a processing gas, so that the mask layer as shown in FIG. The microlens 100 is formed by transferring the lens shape 103 to the lens material layer 102.

  Here, the mask layer 103 is patterned by a photolithography process and formed into a lens shape, but is softened by a heat treatment after the exposure process. For this reason, when the lenses are provided close to each other, the lenses are brought into contact with each other by the surface tension due to the softening, and the lens shape is destroyed. Therefore, in the mask layer 103, the lenses are arranged at an interval of about 0.2 to 0.5 μm, for example, as D1, so that the lenses do not come into contact with each other. The interval is about. Therefore, intervals corresponding to D1 and D2 are also formed between the microlenses 100 transferred to the lens material layer 102.

  However, when the lens material layer 102 is made of an inorganic material, the distances D1 and D2 between the microlenses 100 transferred to the lens material layer 102 are formed on the mask layer as shown by the distance D1 in FIG. There is a problem that it becomes larger than the formed distances d1 and d2 (hereinafter referred to as initial distances d1 and d2).

Here, for example, in a method of forming a microlens using a silicon nitride film as a lens material, the technique of Patent Document 1 has been reported as a technique for reducing the distance between microlenses. In this technique, SF ₆ gas and CHF ₃ gas are used as processing gases, and the flow rate ratio of these gases is adjusted to etch two layers of a mask material layer and a lens material layer made of a Si ₃ N ₄ film. Thus, deposits are deposited on the side walls of the lens formed in the mask layer to reduce the distance between the lenses, and the distance between the microlenses is reduced by transferring the deposit.

  However, according to the verification by the present inventors, it is recognized that the distances D1 and D2 cannot be sufficiently narrowed even by the method of this document, and it is insufficient to solve the problem of the present invention. I can say that. This is one of the factors that hinder the improvement of sensitivity in solid-state imaging devices using microlenses made of inorganic materials. For this reason, organic and inorganic materials can be freely used as materials for microlenses. It was not possible to secure a sufficient degree of freedom in selecting the material.

JP 2005-101232 A

The present invention has been made under such circumstances, and an object of the present invention is to provide a microlens that can control the lens shape, thereby increasing the surface area and reducing the interval between adjacent microlenses. It is to provide a forming method .

Therefore, the method for forming the microlens of the present invention is
Forming a lens material layer made of an inorganic material on a substrate;
Next, an intermediate layer made of an organic material is formed on the lens material layer,
Next, a step of forming a mask layer made of an organic material on the lens material layer,
Next, forming a lens shape on the mask layer;
Next, the process gas containing carbon and fluorine is turned into plasma, and the mask layer and the intermediate layer are etched to transfer the lens shape of the mask layer to the intermediate layer;
Next, the processing gas containing SF ₆ gas and CHF ₃ gas is turned into plasma, and the intermediate layer and the lens material layer are etched to transfer the lens shape of the intermediate layer to the lens material layer. Forming the step.

  The lens material layer is formed of a film selected from a silicon nitride film, a silicon oxide film, and a silicon oxynitride film. In the step of etching the mask layer and the intermediate layer, a gas containing carbon and fluorine is processed. Used as gas. The mask layer may be formed of a resist film or a film made of the same kind of organic material as the intermediate layer.

When the lens material layer is a silicon nitride film, the etching step of etching the intermediate layer and the lens material layer is performed by dividing an etching rate of the lens material layer by an etching rate of the intermediate layer. Is preferably performed under an etching condition in which the etching selectivity is 1.0 or more and 1.6 or less, and more preferably in an etching condition in which the etching selectivity is 1.4 or more and 1.6 or less. When the lens material layer is a silicon oxide film, the etching step of etching the intermediate layer and the lens material layer is performed by dividing an etching rate of the lens material layer by an etching rate of the intermediate layer. Is preferably performed under an etching condition where the etching selectivity is 1.7 or more, and more preferably, the etching condition is such that the etching selectivity is 1.8 or more. Here, the etching selectivity is controlled by adjusting the flow rate ratio of SF ₆ gas and CHF ₃ gas, for example.

The microlens can be used as a condensing microlens provided so as to correspond to each of a plurality of photosensitive portions arranged in a matrix in a solid-state imaging device .

  According to the present invention, as will be apparent from the examples described later, the lens shape can be controlled, whereby a microlens having a large surface area can be formed, and the interval between adjacent microlenses is reduced. be able to.

First, an example of a method for forming a microlens according to the present invention will be described using a method for manufacturing a CCD solid-state imaging device having a microlens as an example. FIG. 1 shows an example of the configuration of the CCD solid-state imaging device. In FIG. 1, reference numeral 2 denotes a semiconductor substrate, for example, a Si substrate, provided with photosensitive portions 21 and vertical registers 22 arranged in a matrix on the surface. Light incident on the photosensitive unit 21 is photoelectrically converted by a photodiode and transferred to an output unit (not shown) by a vertical register 22. A conductive film 23 made of, for example, polysilicon and serving as a transfer electrode is provided in a region other than the photosensitive portion 21 on the upper layer side of the Si substrate 2, and a light-shielding made of, for example, aluminum is provided in the upper region of the conductive film 23. A film 24 is formed.

  The light shielding film 24 is for suppressing the incidence of light on the conductive film 23 while making the light incident on the photosensitive portion 21. For this reason, the region corresponding to the photosensitive portion 21 of the light shielding film 24 has no light. An opening for allowing the light to enter is formed. On the light shielding film 24, a planarizing film 25 made of, for example, a polyimide or polystyrene resin is formed.

  A color filter layer 26 is formed on the planarizing film 25, and microlenses 3 made of an inorganic material are formed on the color filter layer 26 in regions corresponding to the photosensitive portions 21. ing. The microlens 3 is for condensing light onto the photosensitive portion 21, and is formed to have a larger planar size than the photosensitive portion 21 in order to collect light over a wider range.

  Next, a method for forming the above-described microlens 3 will be described with reference to FIGS. 2 and 3. As described above, the microlens 3 is formed in a matrix on the wafer W forming the substrate, and X, Y An interval D1 is formed between the microlenses 3 adjacent in the direction, and an interval D2 is formed between the microlenses 3 adjacent in the oblique direction (see FIG. 16). The present invention aims to make the distance D1 and the distance D2 smaller than the initial distances d1 and d2 formed in the mask layer 33 by adjusting the lens shape, but by narrowing the distance D1, Since the interval D2 can also be automatically narrowed, the following description will be given focusing on the interval D1.

  First, after the photosensitive portion 21 and the vertical register 22 are formed on the Si substrate 2, the conductive film 23 and the light shielding film 24 are formed, and then the planarization film 25 and the color filter layer 26 are formed in this order. As shown in FIG. 1, a lens material layer 31 made of an inorganic material such as a silicon nitride film is formed on the color filter layer 26 with a thickness of about 1 μm, for example, and an intermediate layer 32 and an upper layer on the lens material layer 31 are formed. The mask layer 33 is formed in this order. The intermediate layer 32 is made of a film made of an organic material, for example, with a thickness of about 0.5 to 1.5 μm, and the mask layer 33 is made of a film made of an organic material, for example, with a thickness of about 0.6 μm. Has been.

Here, the silicon nitride film (silicon nitride film) is a film containing silicon (Si) and nitrogen (N), and it is assumed that the main component is a Si ₃ N ₄ film. This will be described as “film”. As an example of the method of forming this SiN film, a gas containing silicon and nitrogen, such as dichlorosilane (SiCl ₂ ) gas and ammonia (NH ₄ ) gas, is used as a raw material gas, and these dichlorosilane gas and ammonia gas are used. By forming into plasma, each active species of silicon and nitrogen contained in the plasma is deposited on the color filter layer 26.

  The organic film forming the intermediate layer 32 is an organic film made of an organic material such as C, H, and O. For example, a phenol resist film, an acrylic resist film, a KrF resist film, a cycloolefin maleic anhydride is used. A resist film (COMA resist film) or the like as a platform can be used. The intermediate layer 32 is formed on the lens material layer 31 by applying a predetermined resist solution by spin coating.

  Further, as the mask layer 33, a phenol-based, acrylic-based resist film such as a KrF-based resist film, an I-line resist film, an X-ray-based resist film, or a resist film (COMA resist film) using cycloolefin maleic anhydride as a platform. Can be used. The mask layer 33 is formed on the intermediate layer 32 by applying a predetermined resist solution by spin coating, and thereafter patterned by a photolithography process and subjected to a heat treatment, so that the predetermined layer as shown in FIG. Processed into a lens shape.

Next, as shown in FIG. 2A, the mask layer 33 and the intermediate layer 32 are etched using a first processing gas containing carbon and fluorine, for example, CF ₄ gas and C ₄ F ₈ gas. The lens shape of the mask layer 33 is transferred to the intermediate layer 32. Here, in this etching process, CF radicals and (CF ₂ ) n radicals and the like are used as etching species with the F radicals in the dissociated products dissociated from these gases formed by the plasma of CF ₄ gas and C ₄ F ₈ gas. It is presumed that the etching proceeds while the etching with F radicals and the deposition with CF radicals are performed simultaneously, acting as deposition species. At this time, since the deposition species is deposited in the peripheral region of the lens shape of the mask layer 33, if a predetermined etching condition is selected, the lens shape of the mask layer 33 is increased in lens width by this deposition. By transferring the layer 33, the intermediate layer 32 can increase the lens width.

  Incidentally, at the initial stage of etching, although the reason is not clear, it is recognized that the interval D1 is larger than the initial interval d1, as indicated by a broken line in FIG. However, the intermediate layer 32 is formed of an organic material containing C, and C contained in the deposition species is generated from the intermediate layer 32 during the etching. Accordingly, the generated C radicals are not hindered by the generated C, but rather promoted, and as the etching progresses, the increased distance D1 is quickly filled with the deposit. It is considered that the spreading speed of the lens shape increases.

  Since the etching and deposition are simultaneously performed in this manner, the lens shape itself of the mask layer 33 is increased, and the deposit is embedded in the gap D1 of the intermediate layer 32, as shown in FIG. As the shape of the intermediate layer 32 increases, the distance D1 is reduced. By selecting the optimum etching conditions, the bottoms of the intermediate layer 32 come into contact with each other, the distance D1 becomes zero, and the distance D2 approaches zero as much as possible.

Next, as shown in FIG. 2B, the intermediate layer 32 and the lens material layer 31 are etched using a second processing gas composed of SF ₆ gas and CHF ₃ gas, whereby the lens shape of the intermediate layer 32 is obtained. Is transferred to the lens material layer 31. Here, in this etching process, SF ₆ gas and CHF ₃ gas are turned into plasma, so that F radicals in dissociated products dissociated from these gases are used as etching species as C radicals, CF radicals, CF ₂ radicals, It is assumed that CF ₃ radicals and the like act as deposition species, respectively, and etching proceeds while etching with F radicals and deposition with C radicals and the like are performed simultaneously.

Here, the CF radical (CF ^* ), CF ₂ radical (CF ₂ ^* ), CF ₃ radical (CF ₃ ^* ), etc. act on the SiN film constituting the lens material layer 31 according to the following reaction. It is considered a thing. In these reaction formulas, SiF ₄ ↑ and N ₂ ↑ indicate that SiF ₄ gas and N ₂ gas are generated, respectively, and C ↓ indicates that C acts as a deposition species on the lens material layer 31.

Si ₃ N ₄ + 12CF ^* → 3SiF ₄ ↑ + 2N ₂ ↑ + 12C ↓
Si ₃ N ₄ + 6CF ₂ ^* → 3SiF ₄ ↑ + 2N ₂ ↑ + 6C ↓
Si ₃ N ₄ + 4CF ₃ ^* → 3SiF ₄ ↑ + 2N ₂ ↑ + 4C ↓
At this time, the deposited species composed of the C radicals and the like are deposited on the lens-shaped peripheral region of the intermediate layer 32, so that the lens width further increases, and the intermediate layer 32 is transferred, whereby the lens material layer 31 is transferred. The lens width increases. On the other hand, at the initial stage of etching, the distance D1 between the lens material layers is larger than the initial distance d1, as described above. Here, in the etching of the lens material layer 31, nitrogen (N ₂ ) gas is generated by reaction with C radicals and the like as shown in the above-described reaction formula, but the deposition of C radicals and the like is inhibited by this N ₂ gas. It is thought that. For this reason, it is presumed that, compared with the etching of the intermediate layer 32 made of an organic film, the embedding by the deposit in the distance D1 of the lens material layer 31 is difficult to proceed, and the spreading speed of the lens shape is reduced.

  However, as described above, the interval D1 of the intermediate layer 32 is sufficiently narrowed, and the deposition species tend to be deposited in the peripheral region of the lens shape of the intermediate layer 32. Since the lens shape of the layer 32 further increases the lens width, when this is transferred, the lens shape of the lens material layer 31 has the distance D1 as shown in FIGS. 2 (c) and 3 (c). It will be narrowed. By selecting the optimum etching conditions in this way, the distance D1 becomes zero, and the microlens 3 is formed in which the distance D2 approaches zero as much as possible.

  Here, in FIGS. 2 and 3, the shape of the microlens 3 is semicircular, but the curvature may be changed depending on the type and configuration of the film so that the planar shape is rectangular. Further, such microlenses 3 are arranged in, for example, a lattice arrangement or a honeycomb arrangement, but the arrangement interval may be the same or different in the X direction and the Y direction. Also good.

  Next, a plasma processing apparatus for forming the microlens 3 will be described with reference to FIG. In the figure, reference numeral 4 denotes an airtight cylindrical processing chamber whose wall portion is made of, for example, aluminum. The processing chamber 4 includes an upper chamber 4A and a lower chamber 4B larger than the upper chamber 4A. The lower chamber 4B is grounded.

  The processing chamber 4 includes a mounting table 41 that also serves as a lower electrode for supporting the wafer W as a substrate substantially horizontally. The mounting table 41 is made of, for example, aluminum. An electrostatic chuck 42 for attracting and holding the wafer W by electrostatic attraction is provided on the surface of the mounting table 41. In the figure, reference numeral 42 a denotes a power supply unit of the electrostatic chuck 42. A focus ring 43 is disposed around the wafer W placed on the surface of the electrostatic chuck 42, and plasma is focused on the wafer W on the placement table 41 through the focus ring 43 when plasma is generated. It is configured. The mounting table 41 is supported by a support base 45 made of a conductor via an insulating plate 44, and the surface of the mounting base 41 is positioned in the lower chamber 4B via the support base 45 by an elevating mechanism including a ball screw mechanism 46, for example. It is configured to be movable up and down between the mounting position and the processing position shown in FIG. In the figure, reference numeral 47 denotes a bellows made of, for example, stainless steel (SUS), and the support base 45 is electrically connected to the processing chamber 4 through the bellows 47.

  Inside the mounting table 41, a refrigerant chamber 48 for allowing the refrigerant to flow is formed, whereby the surface of the mounting table 41 is controlled to about 40 ° C., for example. The wafer W is controlled to a predetermined temperature, for example, about 60 ° C. by heat input. Further, a gas flow path 49 is provided inside the mounting table 41, and a backside gas that forms a cooling gas is supplied between the electrostatic chuck 42 and the back surface of the wafer W so as to adjust the temperature of the wafer W. It is configured.

A region facing the mounting table 41 in the top wall portion of the processing chamber 4 is configured as a gas supply chamber 5 that also serves as an upper electrode. A large number of gas discharge holes 5a are formed in the lower surface of the gas supply chamber 5, and, for example, CF ₄ gas is used as a first process gas source through a gas supply path 51 serving as a gas supply means on the upper surface. The source 52A and the C ₄ F ₈ gas source 52B are connected, and for example, an SF ₆ gas source 52C and a CHF ₃ gas source 52D are connected as the second processing gas sources, respectively. In the figure, MA, MB, MC, and MD are mass flow controllers, and VA, VB, VC, and VD are valves, and the flow rate adjusting means 50 is constituted by these. In this way, the first processing gas or the second processing gas is supplied almost uniformly over the entire surface of the mounting table 41 from the gas discharge hole 5 a toward the mounting table 41 through the gas supply chamber 5. It has become so.

  Around the upper chamber 4A of the processing chamber 4, a dipole ring magnet 61 having a plurality of anisotropic segmented columnar magnets forming magnetic field forming means is disposed, and a predetermined magnetic field, for example, 100G is applied to the upper chamber 4A. It can be added. Furthermore, a high frequency power supply unit 63 that constitutes a high frequency supply means for plasma formation is connected to the mounting table 41 via a matching unit 62, and a high frequency power of a predetermined frequency, for example, 13.56 MHz, is connected from the high frequency power supply unit 63. Is supplied to the mounting table 41. In this way, the gas supply chamber 5 and the mounting table 41 function as a pair of electrodes, and a high frequency can be generated between the gas supply chamber 5 and the mounting table 41 to convert the processing gas into plasma. The inside of the processing chamber 4 is evacuated to a predetermined degree of vacuum by the vacuum evacuation means 54 via the pressure adjustment means 54A and the exhaust passage 53. In the figure, 55 is a wafer carry-in / out port, and 56 is a gate valve for opening and closing the carry-in / out port 55.

  Further, the plasma processing apparatus 10 is provided with a control unit 57 made of, for example, a computer that serves as a control means. The control unit 57 includes a data processing unit made up of a program, a memory, a CPU, and the like. The control unit 57 sends a control signal to each part of the plasma processing apparatus 10 such as the flow rate adjusting means 50 and the pressure adjusting means 54A, and a command is incorporated to perform plasma processing on the wafer W. Further, for example, the memory has an area in which values of processing parameters such as processing pressure, processing time, gas flow rate, and power value are written, and when the CPU executes each instruction of the program, these processing parameters are read out. Then, a control signal corresponding to the parameter value is sent to each part of the plasma processing apparatus 10. This program (including programs relating to processing parameter input operations and display) is stored in a storage unit 58 such as a computer storage medium such as a flexible disk, a compact disk, or an MO (magneto-optical disk) and installed in the control unit 57.

Next, an etching process performed in the plasma processing apparatus 10 will be described. First, a gate valve (not shown) is opened, and a wafer W having the structure shown in FIG. 1 on the surface thereof is loaded into the processing chamber 4 from the loading / unloading port 55 via a wafer transfer unit (not shown). Delivered on the table 41. Then, the mounting table 41 is raised to the processing position, and the inside of the processing chamber 4 is evacuated to a predetermined degree of vacuum, for example, 5.3 Pa (40 mTorr) through the pressure adjusting unit 54A by the vacuum exhaust unit 54. Next, CF ₄ gas and C ₄ F ₈ gas as the first processing gas are introduced from the gas supply chamber 5 at a flow rate of, for example, 100 sccm and 30 sccm, respectively.

  On the other hand, a high frequency of a predetermined frequency, for example, 13.56 MHz, is supplied to the mounting table 41 from the high frequency power supply unit 63 with a power of, for example, 1400 W. As a result, a high-frequency electric field is formed between the gas supply chamber 5 as the upper electrode and the mounting table 41 as the lower electrode. Here, in the upper chamber 4A, since a horizontal magnetic field is formed by the dipole ring 61, an orthogonal electromagnetic field is formed in the processing space in which the wafer W exists, and a magnetron discharge is generated by the drift of electrons generated thereby. The The first process gas is turned into plasma by the magnetron discharge, and the mask layer 33 and the intermediate layer 32 on the wafer W are etched by the plasma as described above.

Next, the introduction of the first processing gas is stopped, and the inside of the processing chamber 4 is evacuated to a predetermined degree of vacuum, for example, 2.65 Pa (20 mTorr) through the pressure adjusting unit 54A by the vacuum exhaust unit 54. Next, SF ₆ gas and CHF ₃ gas, which are second process gases, are introduced from the gas supply chamber 5 at flow rates of, for example, 30 sccm and ₆₀ sccm, respectively.

  On the other hand, a high frequency of a predetermined frequency, for example, 13.56 MHz, is supplied to the mounting table 41 from the high frequency power supply unit 63 with, for example, 400 W of power. As a result, as described above, magnetron discharge is generated in the processing space where the wafer W exists. Then, the second processing gas is turned into plasma by this magnetron discharge, and the intermediate layer 32 and the lens material layer 31 on the wafer W are etched by the plasma as described above. The wafer W having the microlens 3 formed on the surface in this manner is carried out of the processing chamber 4 through the carry-in / out port 55 by a wafer carrying unit (not shown).

  As described above, in the above-described embodiment, the intermediate layer 32 is provided between the mask layer 33 and the lens material layer 31, and the intermediate layer 32 made of an organic material is first etched using the mask layer 33 under predetermined conditions. Thus, after the lens shape of the intermediate layer 32 is enlarged, the lens material layer 31 made of an inorganic material is etched using the intermediate layer 32 as a mask. Therefore, the shape of the intermediate layer 32 having a lens shape larger than that of the mask layer 33 is transferred to the lens material layer 31, and thus the microlens 3 having a lens shape larger than that of the mask layer 33 can be formed. . Thereby, the interval D1 of the microlens 3 can be made narrower than the initial interval d1, and if the etching conditions are selected, the microlens 3 is formed in which the interval D1 is zero and the interval D2 approaches zero as much as possible. Can do.

Here, it is assumed that the intermediate layer 32 is not provided, the mask layer 33 and the lens material layer 31 made of an SiN film are stacked, and the microlens 3 is formed using a processing gas containing SF ₆ gas and CHF ₃ gas. Considering the case, as described above, in the etching of the SiN film, the presence of N ₂ gas hinders the deposition of C radicals and the like, and the film forming property is smaller than the etching of the organic film. It is difficult for the deposit to be embedded in the gap D1 of the lens material layer 31.

  At this time, it is conceivable to increase the film formability by performing the etching process for a long time and to advance the filling of the deposit into the gap D1, but the SiN film has a high in-plane uniformity. In order to ensure the thickness, the film thickness of about 1 μm is the limit, and there is a background that the film thickness cannot be increased any more, and the etching time cannot be made long. Since the film formability cannot be increased in such a limited film thickness, it is assumed that it is difficult to narrow the distance D1 between the microlenses 3 than the initial distance d1 between the mask layers 33.

  Here, in the etching of the lens material layer 31, the etching selection ratio of the lens material layer 31 to the intermediate layer 32 ((etching speed of the lens material layer 31) / (etching speed of the intermediate layer 32)): By controlling the ratio, it is possible to control the lens shape of the microlens 3 as will be apparent from the examples described later.

At this time, the etching selectivity can be controlled by adjusting the flow ratio of SF ₆ gas to CHF ₃ gas. That is, in the etching of the lens material layer 31, as described above, the F radical in the dissociated product dissociated from the SF ₆ gas and the CHF ₃ gas acts as an etching species, and the C radical acts as a deposition species. By adjusting the amount of F radicals and the amount of C radicals, etc., the etching property and the deposition property are adjusted, whereby the etching selectivity can be controlled.

  Further, from the examples described later, when the etching selectivity is small, the deposition property with respect to the etching property becomes small, while when the etching selectivity ratio is large, the deposition property with respect to the etching property becomes large, and the lens shape becomes large. If the etching selectivity becomes too large, the deposition property with respect to the etching property becomes too large, the etching rate is lowered and an etch stop occurs, and the etching selectivity becomes uniform in the etching rate within the wafer surface. Therefore, it is necessary to obtain an appropriate range of the etching selection ratio based on these facts.

  Therefore, in order to form the microlens 3 in a state in which the lens shape is controlled and the in-plane uniformity of the lens shape is enhanced within the processing time considering the throughput of the production line, the etching selectivity is 1. It is preferable to perform the etching process under an etching condition of 0 or more and 1.6 or less. Especially when the etching selectivity is in the range of 1.4 or more and 1.6 or less, it is the same as or smaller than the initial interval d1. The microlens 3 having the interval D1 can be formed, and the microlens having the interval D1 of zero and the interval D2 approaching zero can be formed by further narrowing the etching selection ratio.

Further, the lens shape can be controlled and the size of the interval D1 can be adjusted by controlling the supply amount of the high frequency power supplied into the processing container 4 and the processing pressure in the processing container 4 as will be described later. This is because the amount of energy given to the SF ₆ gas and the CHF ₃ gas varies depending on the supply amount of the high-frequency power and the change in the processing pressure, and thereby the dissociated product dissociated from the SF ₆ gas and the CHF ₃ gas. Since the generation amount of F radicals, C radicals, etc. is different, even if the flow rate ratio of SF ₆ gas and CHF ₃ gas is the same, the amount of F radicals contributing to etching, C radicals contributing to deposition, etc. It is inferred that the amount changes. Therefore, it is desirable to perform the etching under the etching conditions in which the etching rate of the lens material layer 31 is higher than the etching rate of the intermediate layer 32 in order to reduce the interval D1, and the etching selectivity, the supply amount of high frequency power, the processing pressure, etc. By adjusting the parameters of the etching process, the adjustment range of the lens shape is increased, and the microlenses 3 in which the distance D1 and the distance D2 are close to zero or zero can be formed.

  As described above, in the present invention, the microlens 3 is formed using the three-layer structure of the mask layer 33 and the intermediate layer 32 made of an organic material, and the lens material layer 31 made of an inorganic material. By selecting the lens shape, the lens shape can be controlled, the lens width is larger than the lens shape of the mask layer 33, and the inter-lens distance (distance D1) between adjacent lenses is as extremely small as about 0 to 0.1 μm. The microlens 3 made of an inorganic material can be formed. In such a microlens 3, since the degree of light collection on the photosensitive portion 21 is large, high sensitivity can be ensured.

  As described above, since the microlens 3 made of an inorganic material can be put into practical use, the degree of freedom of material selection in which the material of the microlens 3 can be freely selected from an organic material and an inorganic material according to a target wavelength region. Is increased. In addition, by providing the microlens 3 formed of different materials on the solid-state imaging device over a plurality of layers, each microlens 3 selectively collects light in each specific wavelength region, and each weak wavelength region. It is predicted that it will be possible to compensate.

In the above, as the first processing gas, a gas selected from CF ₄ gas, SF ₆ gas, C ₂ F ₆ gas, and C ₃ F ₈ gas, C ₄ F ₈ gas, C ₅ F ₈ gas, and C _{4 are used.} A gas combined with a gas selected from F ₆ gas, C ₂ F ₆ gas, and C ₃ F ₈ gas can be used. As the second processing gas, SF ₆ gas and CHF ₃ gas may be combined with oxygen (O ₂ ) gas.

  The intermediate layer 32 and the mask layer 33 are both made of an organic material, but they may be made of the same type of film or different types of films. Also good. When these are formed of the same kind of film, the mask layer 33 and the intermediate layer 32 are formed of, for example, a phenol resist film, an acrylic resist film, a KrF resist film, a resist film (COMA) using cycloolefin maleic anhydride as a platform. Resist film). In this case, since the etching selectivity of the mask layer 33 and the intermediate layer 32 is the same, the shape of the mask layer 33 is transferred as it is to the intermediate layer 32, which is effective in that the lens shape can be easily controlled. .

  Further, the intermediate layer 32 may be formed of a plurality of layers of one or more layers, and these may be formed of the same kind of organic film or different kinds of organic films. By providing a plurality of intermediate layers 32 in this manner, the width of adjustment of the lens shape of the intermediate layer 32 is increased, and the width of adjustment of the lens shape of the microlens 3 is also increased by transferring this. .

Furthermore, as an inorganic material for forming the lens material layer 31, a silicon oxide film, a silicon oxynitride film, or the like can be used. Here, a case where a silicon oxide film is used as the lens material layer 31 will be described. This silicon oxide film is a film containing silicon and oxygen (O), and is generally known as a silicon dioxide film (SiO ₂ film), and will be described here as an SiO ₂ film. First and an example of a method for forming the SiO ₂ film, as the raw material gas for forming the SiO ₂ film, for example, tetraethyl orthosilicate _{(Si (OC 2 H 5)} 4) organic source, such as steam (gas) And an oxygen gas are used, and the tetraethylorthosilicate gas and the oxygen gas are converted into plasma, so that the SiO ₂ film is formed on the color filter layer 26 by, for example, 4 μm by the active species of silicon and oxygen contained in the plasma. The film thickness is formed.

In the etching process of the lens material layer 31 made of the SiO ₂ film, as in the etching of the SiN film, the SF radicals in the dissociated products dissociated from these gases are obtained by converting SF ₆ gas and CHF ₃ into plasma. As etching species, C radicals, CF radicals, CF ₂ radicals, CF ₃ radicals, etc. act as deposition species, respectively, and etching progresses while etching by F radicals and deposition by C radicals are performed simultaneously. It is guessed.

At this time, it is considered that CF radical (CF ^* ), CF ₂ radical (CF ₂ ^* ), CF ₃ radical (CF ₃ ^* ) and the like act on the SiO ₂ film according to the following reaction.

3 / 4SiO ₂ + CF ₃ ^* → 3/4 SiF ₄ + CO + 1 / 2O
1 / 2SiO ₂ + CF ₂ ^* → 1 / 2SiF ₄ + CO
1 / 4SiO ₂ + CF ^* → 1 / 4SiF ₄ + 1 / 2CO + 1 / 2C
As described above, in the etching of the SiO ₂ film, O and CO are generated and C is released as a deposition component. Due to the influence of O and CO, the film is formed as compared with the case where the intermediate layer 32 which is an organic film is etched. It is considered that these O and CO are less likely to inhibit the deposition of C radicals and the like than N ₂ gas generated during etching of the SiN film. Further, as will be apparent from the examples described later, the etching selectivity does not increase and the phenomenon that the etching does not proceed does not occur, so that the lens shape can be formed larger as the etching selectivity increases. Expected.

For this reason, when the lens material layer 31 is a SiO ₂ film, by performing an etching process under an etching condition in which the etching selection ratio is 1.7 or more, within a processing time considering the throughput of the production line, The microlens 3 can be formed in a state where the lens shape can be controlled within a desired range, and by selecting an etching condition, the interval D1 is narrower than the initial interval d1, the interval D1 is zero, and the interval D2 is infinitely zero. A microlens approaching can be formed. Further, when etching is performed with this etching selectivity, the in-plane uniformity of the etching rate is good, as will be apparent from the examples described later.

As described above, when the lens material layer 31 is a SiN film or a SiO ₂ film, the microlenses 3 can be formed by selecting the etching conditions so that the distances D1 and D2 are zero or close to zero. It is assumed that the same effect can be obtained when the microlens 3 is formed using a silicon oxynitride film as a material. This silicon oxynitride film is a film containing silicon, nitrogen, and oxygen, and is a SiON film here. For example, this SiON film uses a processing gas containing silicon, nitrogen, and oxygen, and plasma CVD ( Chemical Vapor Deposition) method.

  The microlens of the present invention can also be applied to a CCD solid-state imaging device having a structure shown in FIGS. 5A and 5B or a microlens 3 formed on a CMOS sensor. FIG. 5A shows an example in which an in-layer microlens 27 is provided in addition to the microlens 3 formed on the surface, and this in-layer microlens 27 has the structure of the color filter layer 26 in the structure shown in FIG. Formed in the lower layer. In the figure, reference numeral 28 denotes a planarizing film formed on the surface of the in-layer microlens 27 (in some cases, only the color filter layer 26), and other structures are the same as those shown in FIG. In such a structure, the surface microlens 3 is formed by the method of the present invention. FIG. 5B shows an example in which the microlens 3 is formed directly on the light shielding film 24 in the structure shown in FIG. 1, and the microlens 3 on the surface is formed by the method of the present invention.

Examples carried out to confirm the effects of the present invention will be described below. In the following experiment, as shown in FIG. 1, a photosensitive portion 21, a vertical register 22, a conductive film 23, and a light shielding film 24 are formed on a Si substrate 2, and a planarizing film 25 and a color filter layer are formed thereon. 26, a wafer W in which a lens material layer 31, an intermediate layer 32, and a mask layer 33 formed in a predetermined lens shape are formed in this order from the lower side was used. As the etching apparatus, the plasma etching apparatus shown in FIG. 4 is used.
1. When the lens material layer 31 is formed of a SiN film (Example 1-1)
As shown in FIG. 6A, an intermediate layer 32 made of a phenol resist film on a lens material layer 31 having a film thickness of 1 μm, and a mask layer made of a phenol resist film and formed in a predetermined lens shape. Etching is performed on the 8-inch wafer W on which the 33 is formed in this order under the following conditions, and scanning is performed for each of the lens shapes of the mask layer 33, the intermediate layer 32, and the lens material layer 31 (microlens 3). The planar shape was imaged using a scanning electron microscope (SEM), and based on this, the distance D1 was measured for each of the mask layer 33, the intermediate layer 32, and the lens material layer 31. FIG. 6A shows a trace of a photograph taken by the SEM (hereinafter referred to as “SEM photograph”) and the distance D1.
[Etching conditions for intermediate layer 32]
Processing gas: CF ₄ / C ₄ F ₈ = 100/30 sccm
High frequency power: 1400W
Processing pressure: 5.3 Pa (40 mTorr)
Setting temperature of mounting table: 0 ° C
Processing time: Etching was performed for 199 seconds by EPD (end-point detector using a plasma emission spectrum analyzer). Here, the end point at the time of etching was detected based on the calculation result such as the ratio of the emission spectrum intensity (wavelength 260 nm) due to the CF radical and the emission spectrum intensity (wavelength 387.2 nm) due to the CN radical, and the etching was stopped.
[Etching Conditions for Lens Material Layer 31]
Processing gas: SF ₆ / CHF ₃ / O ₃ = 60/50/25 sccm
Etching selectivity: 0.95
High frequency power supply: 400W
Processing pressure: 2.65 Pa (20 mTorr)
Setting temperature of mounting table: 0 ° C
Processing time: The etching was stopped until the lens material layer 31 was etched to 750 nm.
(Comparative Example 1)
As shown in FIG. 6B, a wafer W is formed on a lens material layer 31 having a film thickness of 1 μm, and a mask layer 33 made of a phenol-based resist film and having a predetermined lens shape is formed in this order. On the other hand, etching is performed under the following conditions, SEM photographs are taken for the planar shapes of the respective lens shapes of the mask layer 33 and the lens material layer 31, and based on this, the interval between the mask layer 33 and the lens material layer 31 is spaced. D1 was measured. FIG. 6B shows a trace of this SEM photograph and the distance D1.
[Etching Conditions for Lens Material Layer 31]
Process gas: SF ₆ / CHF ₃ = 60/60 sccm
Etching selectivity: 1.09
High frequency power supply: 400W
Processing pressure: 2.65 Pa (20 mTorr)
Setting temperature of mounting table: 40 ° C
Processing time: The etching was stopped until the lens material layer 31 was etched to 750 nm.
(Experimental result)
Focusing on the distance D1 (d1), in Example 1, the mask layer 33 is 320 nm, the intermediate layer 32 is 100 nm, and the microlens 3 is 358 nm. In Comparative Example 1, the mask layer 33 is 500 nm, and the microlens 3 is 700 nm. As a result, the distance D1 is about 1.1 times larger in the microlens 3 than in the mask layer 33 in the first embodiment, whereas the microlens 3 is larger in the comparative example 1 than in the mask layer 33. Then, it was recognized that it spreads about 1.4 times.

Here, the etching selectivity is 0.95 in Example 1 and 1.09 in Comparative Example 1. Comparative Example 1 is larger, and the larger the etching selectivity is, the stronger the deposition and the larger the lens shape. However, it is still recognized that the lens shape of Example 1 can be made larger and the distance D1 can be narrowed, and thus the effectiveness of the present invention has been understood. It was also confirmed that the distance D1 between the intermediate layers 32 of Example 1 was narrower than the distance D1 between the mask layers 33.
(Example 1-2: Control of lens shape by adjusting etching selectivity)
Etching is performed on the same wafer W as in Example 1-1 with respect to the lens material layer 31 while changing the etching selection ratio in the range of 0.95 to 1.75, and the mask layer 33, the intermediate layer 32, and the lens. For each of the material layers 31, SEM photographs were taken for the planar shape and the cross-sectional shape, the change in the lens shape was observed, and the distance D1 (d1) and the etching depth were measured for each based on the SEM photograph.

Here, the etching depth (etching amount) is an index of the etching amount of the lens material layer (SiN film) 31, and the thickness of the lens material layer 31 after the etching of the intermediate layer 32 shown in FIG. It is calculated from the difference (X−Y) between the thickness X and the thickness Y of the lens material layer 31 after the etching of the lens material layer 31 shown in FIG. At this time, the thicknesses X and Y are thicknesses of regions where no lens shape is formed. In this embodiment, the etching of the intermediate layer 32 is performed by detecting the etching end point by the plasma emission spectrum, and when the etching of the intermediate layer 32 is completed, the surface portion of the lens material layer 31 may be slightly etched. FIG. 7A shows a state where the surface of the lens material layer 31 is etched. Further, since the interval D1 between the intermediate layer 32 and the lens material layer 31 may not be zero depending on the etching conditions, a state in which these are provided with a predetermined interval D1 is shown here. A trace of this SEM photograph, the distance D1 and the etching depth are shown together in FIG. Further, FIG. 9 shows the relationship between the etching selectivity and the interval D1.
[Etching conditions for intermediate layer 32]
It carried out on the same conditions as Example 1-1.
[Etching Conditions for Lens Material Layer 31]
Process gas: Separately Etching selection ratio: Separately High-frequency power supply: 400W
Processing pressure: 2.65 Pa (20 mTorr)
Setting temperature of mounting table: 0 ° C
Processing time: The etching was stopped until the lens material layer 31 was etched to 750 nm.

  The etching selectivity was controlled by changing the flow rate ratio of the processing gas. The relationship between the etching selectivity and the flow rate ratio of the processing gas is as follows.

Selection ratio 0.95: SF ₆ / CHF ₃ / O ₂ = 60/50/25 sccm
Selection ratio 1.42: SF ₆ / CHF ₃ = 30/60 sccm
Selection ratio 1.59: SF ₆ / CHF ₃ = 28/60 sccm
Selection ratio 1.66: SF ₆ / CHF ₃ = 29/60 sccm
Selection ratio 1.75: SF ₆ / CHF ₃ = 25/60 sccm
It was recognized that by adjusting the etching selectivity according to FIGS. 8 and 9, the lens shape changes and the distance D1 can be controlled. As a result, when the etching selection ratio is 0.95, the interval D1 becomes larger than the initial interval d1, but the interval D1 decreases as the etching selection ratio increases, while the etching selection ratio is 1. When it was .66 or more, it was confirmed that the etching depth of the lens material layer 31 did not reach the target value of around 750 nm, and a phenomenon in which the etching did not proceed occurred. If the etching selectivity becomes too large, the etching does not proceed because the etching by the F radicals also proceeds, but the deposition by the C radicals or the like further proceeds, so the ratio of the deposition amount to the etching amount. It is assumed that this is because the etching stop becomes too high.

Thus, in order to form the microlens 3 having the distance D1 narrower than the initial distance d1 while securing a certain etching amount in combination with the data of FIG. 9, the etching selectivity is 1.0 or more. It is preferable to perform the etching of the intermediate layer 32 and the microlens 3 under the condition of 1.6 or less. In particular, if the etching selectivity is in the range of 1.4 to 1.6, the distance D1 is smaller than 150 nm. Thus, it is understood that the distance D1 between the microlenses 3 can be formed to be the same as or narrower than that of the intermediate layer 32, which is more preferable.
(Example 1-3: Relationship between etching selectivity and in-plane uniformity of etching rate)
The lens material layer 31 is etched by changing the etching selectivity in the range of 0.86 to 3.25 with respect to the wafer W similar to the example 1-1, and the etching rate and the etching rate of the lens material layer 31 are etched. The in-plane uniformity was measured. The etching rate indicates an average value of etching rates measured at 25 points on the wafer surface, and the in-plane uniformity of etching rate is obtained by etching a deviation of etching rates measured at 25 points on the wafer surface. The value divided by the absolute value of the speed is shown, and the closer this value is to zero, the higher the in-plane uniformity of the etching speed. The etching conditions for the intermediate layer 32 and the lens material layer 31 are as follows.
[Etching conditions for intermediate layer 32]
It carried out on the same conditions as Example 1-1.
[Etching Conditions for Lens Material Layer 31]
Process gas: Separately Etching selection ratio: Separately High-frequency power supply: 400W
Processing pressure: 2.65 Pa (20 mTorr)
Setting temperature of mounting table: 0 ° C
Processing time: The etching was stopped until the lens material layer 31 was etched to 750 nm.

  The etching selectivity was controlled by changing the flow rate ratio of the processing gas. The relationship between the etching selectivity and the flow rate ratio of the processing gas is as follows.

Selection ratio 0.86: SF ₆ / CHF ₃ / O ₂ = 60/25/30 sccm
Selection ratio 0.95: SF ₆ / CHF ₃ / O ₂ = 60/50/25 sccm
Selection ratio 1.42: SF ₆ / CHF ₃ = 30/60 sccm
Selection ratio 1.59: SF ₆ / CHF ₃ = 28/60 sccm
Selection ratio 1.66: SF ₆ / CHF ₃ = 29/60 sccm
Selection ratio 1.75: SF ₆ / CHF ₃ = 25/60 sccm
Selection ratio 2.17: SF ₆ / CHF ₃ = 20/60 sccm
Selection ratio 3.25: SF ₆ / CHF ₃ = 15/60 sccm
FIG. 10 shows the relationship between the flow rate ratio of the processing gas and the etching selectivity, the etching rate, and the in-plane uniformity of the etching rate. As a result, when the etching selection ratio is 1.75 or more, it is recognized that the in-plane uniformity of the etching rate is abruptly deteriorated, and the intermediate layer 32 under the condition that the etching selection ratio is 1.0 to 1.6. It was confirmed that the in-plane uniformity of the lens shape can be secured by etching the microlenses 3.
2. When the lens material layer 31 is formed of a SiO ₂ film (Example 2-1: Control of the lens shape by adjusting the etching selectivity)
On the lens material layer 31 having a thickness of 4.2 μm, an intermediate layer 32 made of a phenol resist film and a mask layer 33 made of a phenol resist film formed in a predetermined lens shape were formed in this order from the bottom. Etching of the lens material layer 31 is performed on a 6-inch wafer W while changing the etching selectivity in the range of 1.63 to 2.06, and each of the mask layer 33, the intermediate layer 32, and the microlens 3 is performed. SEM photographs were taken for the planar shape and the cross-sectional shape, the change in the lens shape was observed, and the distance D1 (d1) was measured for each based on this SEM shape. The trace of this SEM photograph and the distance D1 are shown together in FIG. FIG. 12 shows the relationship between the etching selectivity and the distance D1.
[Etching conditions for intermediate layer 32]
Processing gas: CF ₄ / C ₄ F ₈ = 100/30 sccm
High-frequency power supply: 1200W
Processing pressure: 5.3 Pa (40 mTorr)
Setting temperature of mounting table: 0 ° C
Processing time: Etching is performed by EPD for 139 seconds, and the end point at the time of etching is detected based on the calculation result of the ratio of the emission spectrum intensity due to CO radicals (wavelength 226 nm) and the emission spectrum intensity due to CF radicals (wavelength 260 nm). Stopped.
[Etching Conditions for Lens Material Layer 31]
Process gas: Separately Etching selection ratio: Separately High-frequency power supply: 400W
Processing pressure: 2.65 Pa (20 mTorr)
Setting temperature of mounting table: 0 ° C
Processing time: The etching was stopped until the lens material layer 31 was etched by 2.8 μm.

  The etching selectivity was controlled by changing the flow rate ratio of the processing gas. The relationship between the etching selectivity and the flow rate ratio of the processing gas is as follows.

Selection ratio 1.63: SF ₆ / CHF ₃ = 12/60 sccm
Selection ratio 1.80: SF ₆ / CHF ₃ = 10/60 sccm
Selection ratio 2.06: SF ₆ / CHF ₃ = 8/60 sccm
It was recognized that adjusting the etching selection ratio according to FIGS. 11 and 12 changes the lens shape and can control the distance D1. As a result, when the etching selection ratio is 1.7 or more, the interval D1 is 500 nm or less, and when the etching selection ratio is 1.8 or more, it is almost the same as the initial interval d1, and the etching selection ratio is increased. Accordingly, it was confirmed that the distance D1 is reduced, and that the etching amount can be secured even when the etching selectivity is increased, unlike the case where the lens material layer 31 is a SiN film. In this way, when the lens material layer 31 is a SiO ₂ film, it is presumed that even if the etching selection ratio increases, the ratio of the deposition amount to the etching amount does not become too high, and etch stop does not occur.

Thus, from the approximate curve of FIG. 12, in order to form the microlens 3 having the distance D1 which is the same as or narrower than the initial distance d1, the intermediate layer 32 and the intermediate layer 32 under the condition that the etching selectivity is 1.8 or more. It is recognized that it is preferable to perform etching of the microlens 3, and when the etching selectivity is 2.2 or more, it is predicted that the distance D1 can be made zero.
(Example 2-2: Relationship between etching selectivity and in-plane uniformity of etching rate)
Etching the lens material layer 31 is performed on the same wafer W as in Example 2-1, while changing the etching selectivity in the range of 1.63 to 2.06. The in-plane uniformity was measured. The etching rate and the in-plane uniformity of the etching rate were measured at 9 points in the wafer surface with respect to the etching rate, and calculated by the same method as in Example 1-3. The etching conditions for the intermediate layer 32 and the lens material layer 31 are as follows.
[Etching conditions for intermediate layer 32]
It carried out on the same conditions as Example 1-1.
[Etching Conditions for Lens Material Layer 31]
Process gas: Separately Etching selection ratio: Separately High-frequency power supply: 400W
Processing pressure: 2.65 Pa (20 mTorr)
Setting temperature of mounting table: 0 ° C
Processing time: The etching was stopped until the lens material layer 31 was etched by 2.8 μm.

  The etching selectivity was controlled by changing the flow rate ratio of the processing gas. The relationship between the etching selectivity and the flow rate ratio of the processing gas is the same as in Example 2-1.

  FIG. 13 shows the relationship between the flow rate ratio of the processing gas and the etching selection ratio, the etching rate, and the in-plane uniformity of the etching rate. From this result, it was confirmed that the in-plane uniformity of the etching rate was good when the etching selectivity was in the range of 1.63 to 2.06.

(Example 2-3: Relationship between distance D1 and high-frequency power)
Etching is performed on the wafer W of Example 2-1, with the etching selection ratio fixed at 1.6, the amount of high-frequency power supplied being changed, and the distance D1 is measured for the obtained microlens 3, and the distance is measured. The dependence of D1 on the high-frequency power, the etching rate of the lens material layer 31, and the in-plane uniformity of the etching rate were measured. The etching rate and in-plane uniformity of the etching rate were measured by the same method as in Example 2-2. Etching conditions are as follows.
[Etching conditions for intermediate layer 32]
It carried out on the same conditions as Example 1-1.
[Etching Conditions for Lens Material Layer 31]
Processing gas: SF ₆ / CHF ₃ = 12/60 sccm
Etching selectivity: 1.6
High frequency power supply: 400W, 800W
Processing pressure: 2.65 Pa (20 mTorr)
Setting temperature of mounting table: 0 ° C
Processing time: The etching was stopped until the lens material layer 31 was etched by 2.8 μm.

The result is shown in FIG. In FIG. 14, the vertical axis indicates the interval D1, and the horizontal axis indicates the amount of high-frequency power supplied. The etching rate when the power supply amount is 400 W is 186.4 nm / min, the in-plane uniformity of the etching rate is ± 4.5%, and the etching rate when the power supply amount is 800 W is 339.6 nm / min. The in-plane uniformity of the etching rate was ± 3.9%. Thereby, by changing the supply amount of the high frequency power, the lens shape can be adjusted, the size of the interval D1 can be controlled, the etching rate and the in-plane uniformity of the etching rate can be adjusted, and the etching selectivity is further increased. In the case of 1.6, it was recognized that the distance D1 was narrower when the power supply amount was 800 W than when it was 400 W, and the in-plane uniformity of the etching rate was improved.
(Example 2-4: Relationship between interval D1 and processing pressure)
For the wafer W of Example 2-1, the etching selectivity was fixed at 1.6, etching was performed while changing the value of the processing pressure, the distance D1 was measured for the obtained microlens 3, and the distance was measured. The dependence of D1 on the processing pressure, the etching rate of the lens material layer 31, and the in-plane uniformity of the etching rate were measured. The etching rate and in-plane uniformity of the etching rate were measured by the same method as in Example 2-2. Etching conditions are as follows.
[Etching conditions for intermediate layer 32]
It carried out on the same conditions as Example 1-1.
[Etching Conditions for Lens Material Layer 31]
Processing gas: SF ₆ / CHF ₃ = 10 sccm / ₆₀ sccm
Etching selectivity: 1.6
High frequency power supply: 800W
Processing pressure: 1.94 Pa (15 mTorr), 2.65 Pa (20 mTorr)
Setting temperature of mounting table: 0 ° C
Processing time: The etching was stopped until the lens material layer 31 was etched by 2.8 μm.

  The result is shown in FIG. In FIG. 15, the vertical axis indicates the interval D1, and the horizontal axis indicates the processing pressure. The etching rate when the processing pressure is 1.94 Pa is 339.6 nm / min, the in-plane uniformity of the etching rate is ± 3.9%, and the etching rate when the processing pressure is 2.65 Pa is 323.0 nm. / Min, the in-plane uniformity of the etching rate was ± 4.3%. Thus, by changing the processing pressure, the lens shape can be adjusted, the size of the distance D1 can be controlled, the etching rate and the in-plane uniformity of the etching rate can be adjusted, and the etching selectivity is 1.6. In this case, it was confirmed that when the processing pressure was 1.94 Pa, the interval D1 was narrowed, and the in-plane uniformity of the etching rate was also improved.

As described above, the lens shape and the in-plane uniformity depend on the supply amount of high-frequency power and the processing pressure. As described above, the amount of F radicals increases as the supply amount of high-frequency power and the processing pressure increase. As a result, the ratio of the amount of F that contributes to etching and the amount of C and the like that contributes to deposition changes, which is presumed to be reflected in the in-plane uniformity of the lens shape and the etching rate. In the case where the lens material layer 31 is a SiN film, no experiment has been performed on the relationship between the distance D1 and the supply amount of high-frequency power and the processing pressure, but the same result as in the case where the lens material layer 31 is a SiO ₂ film. Is expected to be obtained.

  In the above, the etching process of the present invention can be carried out not only in the above-described plasma processing apparatus but also in an apparatus that generates plasma by other methods. Furthermore, the present invention can be applied not only to the CCD solid-state imaging device but also to the formation of microlenses used in MOS solid-state imaging devices and liquid crystal display devices. Furthermore, the method of the present invention is effective for forming not only the outermost surface microlens but also the in-layer lens. As a substrate on which the microlens of the present invention is formed, a glass substrate is used in addition to the semiconductor wafer. Also good.

It is sectional drawing which shows an example of the CCD solid-state image sensor provided with the micro lens of this invention. It is process drawing which shows the formation method of the said micro lens. It is process drawing which shows the formation method of the said micro lens. It is sectional drawing which shows an example of the magnetron RIE plasma etching apparatus for implementing the etching process for forming the said micro lens. It is sectional drawing which shows the other example of the CCD solid-state image sensor provided with the micro lens of this invention. It is a characteristic view which shows the planar shape of the microlens which shows the result of Example 1-1, and the space | interval D1. It is sectional drawing for demonstrating the etching depth. It is a characteristic view which shows the planar shape and cross-sectional shape of the microlens which show the result of Example 1-2, the space | interval D1, and the etching depth. It is a characteristic view which shows the relationship between the space | interval D1 which shows the result of Example 1-2, and an etching selectivity. It is a characteristic view which shows the etching selectivity which shows the result of Example 1-3, the etching rate, and the in-plane uniformity of an etching rate. It is a characteristic view which shows the planar shape, cross-sectional shape, and space | interval D1 of the microlens which show the result of Example 2-1. It is a characteristic view which shows the relationship between the space | interval D1 which shows the result of Example 2-1, and etching selectivity. It is a characteristic view which shows the etching selection ratio which shows the result of Example 2-2, the etching rate, and the in-plane uniformity of an etching rate. It is a characteristic view which shows the relationship between the space | interval D1 which shows the result of Example 2-3, and the supply amount of high frequency electric power. It is a characteristic view which shows the relationship between the space | interval D1 which shows the result of Example 2-4, and process pressure. It is a top view which shows the formation method of the conventional microlens. It is sectional drawing which shows the formation method of the conventional microlens.

Explanation of symbols

21 Photosensitive part 22 Vertical register 23 Conductive film 24 Light shielding film 25 Flattening film 26 Color filter layer 3 Micro lens 31 Lens material layer 32 Intermediate layer 33 Mask layer 4 Processing chamber 41 Mounting table 42 Electrostatic chuck 5 Gas supply chamber 50 Flow rate adjustment Means 52A CF ₄ gas source 52B C ₄ F ₈ gas source 52C SF ₆ gas source 52D CHF ₃ gas source 54 Vacuum exhaust means 54A Pressure adjusting means 61 Dipole ring magnet 63 High frequency power supply

Claims (10)

Forming a lens material layer made of an inorganic material on a substrate;
Next, an intermediate layer made of an organic material is formed on the lens material layer,
Next, a step of forming a mask layer made of an organic material on the lens material layer,
Next, forming a lens shape on the mask layer;
Next, the process gas containing carbon and fluorine is turned into plasma, and the mask layer and the intermediate layer are etched to transfer the lens shape of the mask layer to the intermediate layer;
Next, the processing gas containing SF ₆ gas and CHF ₃ gas is turned into plasma, and the intermediate layer and the lens material layer are etched to transfer the lens shape of the intermediate layer to the lens material layer. Forming a microlens, comprising: a step of forming a microlens.   2. The method for forming a microlens according to claim 1, wherein the lens material layer is formed of a film selected from a silicon nitride film, a silicon oxide film, and a silicon nitride oxide film. The mask layer is formed a microlens according to claim 1 or 2, characterized in that it is formed from the resist film. The mask layer is formed a microlens according to claim 1 or 2, characterized in that it is formed from consisting of the same type of organic material and the intermediate layer film. When the lens material layer is a silicon nitride film,
The step of etching the intermediate layer and the lens material layer includes an etching condition in which an etching selectivity obtained by dividing the etching rate of the lens material layer by the etching rate of the intermediate layer is 1.0 or more and 1.6 or less. forming a microlens according to any one of claims 1 to 4 carried out be characterized by. The step of etching the intermediate layer and the lens material layer is performed under an etching condition in which an etching selectivity obtained by dividing the etching rate of the lens material layer by the etching rate of the intermediate layer is 1.4 or more and 1.6 or less. The method for forming a microlens according to claim 5, wherein When the lens material layer is a silicon oxide film,
The step of etching the intermediate layer and the lens material layer is performed under an etching condition in which an etching selectivity obtained by dividing the etching rate of the lens material layer by the etching rate of the intermediate layer is 1.7 or more. forming a microlens according to any one of claims 1 to 4, characterized in. The step of etching the intermediate layer and the lens material layer is performed under an etching condition in which an etching selectivity obtained by dividing the etching rate of the lens material layer by the etching rate of the intermediate layer is 1.8 or more. The method for forming a microlens according to claim 7 . The method for forming a microlens according to claim 5 , wherein the etching selectivity is controlled by adjusting a flow rate ratio of SF ₆ gas and CHF ₃ gas. The microlenses in the solid-state imaging device, any one of claims 1, characterized in that to each of the plurality of photosensitive portions arranged in a matrix is a microlens for condensing light is provided so as to correspond 9 A method for forming a microlens as described in 1. above.

Images